Epithelia act as barriers to the movement of small molecules, includingions and drugs, between body compartments. Ions cross the epithelial apical and basolateral membranes via tightly regulated ion-specific transporters and channels. However, exogenously supplied drugs, particularly hydrophilic or charged compounds, are not transported to any appreciable extent across the epithelium. Our labooratory has developed two different classes of peptides that promote selective ion transport or modulate the epithelial barrier that precludes drug access. The peptide sequences are derived from the pore-forming M2 transmembrane (TM) segment of the spinal cord glycine receptor (M2GlyR). Peptides insert and assemble to form transmembrane structures that display distinct, sequence dependent, physiological responses. They either selectively increase chloride transport or form non-selective channels that transiently facilitate the paracellular movement of larger hydrophilic compounds to the basolateral compartment. The former sequences have been proposed as a possible therapeutic intervention for channelopathies such as cystic fibrosis (CF).The latter are currently under investigation for facilitated transport of hydrophilic antibiotics across the cornea. While much is understood about the channel properties of these sequences in their native context little is understood about the assembled membrane structure(s) of the synthetic peptides. Multi-dimensional solution NMR has been used to investigate the structure of peptide monomers in SDS-micelles and modeling of helical bundles has been attempted. We propose to determine the number of transmembrane segments that assemble to form the active structures in model membranes and cells using both chemical and biophysical methods. Recent studies have indicated that different membrane lipid compositions can alter the seconday structure of these peptides. Studies will be conducted using model membranes with compositions that preseve the helical structure of the peptides. We hypothesize that different sized membrane assemblies could be stabilized in different lipid environments. Knowledge of the the number of segments that assemble under physiological conditions will greatly assist future modeling efforts. Additional experiments are planned that will examine the positioning and hydrogen bonding strength of a ring of threonines at position 17 in the sequences that form anion selective channel pores. Various amino acid substitutions in this position will assess the geometry and hydrogen bonding strengths required to optimize a synthetic channel with regard to selectivity and high throughput. These results will be incorporated into future channel designs with the ultimate goal of defining therapeutic structures.